System Doesn't Boot Part 2

Mark Olbert molbert at iterobiopharm.com
Sat Jan 26 18:19:28 PST 2008


For fun, I modified the udev script to recreate /dev/console and /dev/null. After doing that, the entire boot process appears to proceed successfully to completion (i.e., all the other daemons and stuff initialize correctly), but I still don't end up with a login prompt. All I get is a blank screen. And, after a few minutes, I get that console warning about init 0-6 respawning too rapidly and having to be suspended for 5 minutes.

BTW, I understand your observations about a 3 year old system being "old", particularly as regards stuff like udev, which has apparently evolved quite rapidly over that time. All I can say is "it was rock solid until yesterday"...and I was actually planning on building a new, replacement system in the very near future. Timing, as they say, is everything :)

So if you have any more ideas on how to get this puppy to boot properly I'd appreciate hearing them. This mailing list has been great so far, I feel like there's just a little more to go...after which I'll start building that new system.

- Mark

----- Original Message ----
From: Mark Olbert <molbert at iterobiopharm.com>
To: LFS Support List <lfs-support at linuxfromscratch.org>
Sent: Saturday, January 26, 2008 5:09:07 PM
Subject: Re: System Doesn't Boot Part 2


Ken,

One additional thing I noticed: if I do a ls -l /dev at the end of the
 LFS udev script, neither console nor null appear; they're not listed as
 devices.

I don't know if this is because they're not in the ramfs that's mounted
 on top of /dev when udev runs, or if it's related to the problem I'm
 experiencing, but I thought I'd pass it along because it looked odd to
 me.

- Mark

----- Original Message ----
From: Ken Moffat <ken at linuxfromscratch.org>
To: LFS Support List <lfs-support at linuxfromscratch.org>
Sent: Saturday, January 26, 2008 3:39:07 PM
Subject: Re: System Doesn't Boot Part 2


On Sat, Jan 26, 2008 at 03:08:18PM -0800, Mark Olbert wrote:
> Some additional info:
> 
> I checked the logs after the last failed/hanging boot, and the system
 log shows most, if not all, of the shell scripts executing (I can tell
 by the messages they log).
> 
> But I never get a login prompt on the screen, and it looks like the
 system hangs.
> 
> I did notice one oddity in the log file, however. init complains
 about not being able to open /dev/console. Why that should be I don't
 know,
 as I never deleted it. The log also contains messages about init (or
 some process related to it) spawning too fast, etc.
> 
 Hmm, /dev/console looks key to this.  At least until recently,
/dev/console and /dev/null had to be the real device nodes (with
recent udev, there might be copies in /lib/udev, I don't remember,
and it's not relevant for you).  I hit this myself when I restore
from backups, or when I copy a system to a different place to fit a
bigger disk.

 You say you haven't changed this, and in a running system we mount
/dev _on_top_of_ these minimal devices, which should protect them.
Anyway, from your "emergency" slackware system take a look at the LFS
/dev/null and /dev/console with 'ls -l'.  Do they exist, and if so
are they the correct device special nodes, with the correct
permissions ?

 I do wonder if you either ran fsck during the failed boot that
started this, and it found damage and deleted files, or if there is
other damage on the disk.  This all sounds worrying, nearly as
worrying as somebody using such an old system (yes, I know there can
be persuasive reasons to use an old system, but it isn't the comon
case - distros provide updates, and migation paths at end of life,
we have to handle that for ourselves).

ĸen
-- 
das eine Mal als Tragödie, das andere Mal als Farce
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page



-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page






More information about the lfs-support mailing list